Russia-Linked APT29 Uses New Malware in Embassy Attacks


Russia-linked cyberespionage group APT29 has been observed staging new malware for attacks likely targeting embassy-related individuals, Recorded Future reports.

Also referred to as Cozy Bear, the Dukes, Nobelium, and Yttrium, APT29 is a Russian advanced persistent threat (APT) group believed to be sponsored by the Russian Foreign Intelligence Service (SVR). It’s also believed to have orchestrated multiple high-profile attacks, including the 2020 SolarWinds attack.

In October 2022, Recorded Future identified new infrastructure and malware that the cyberespionage group likely set up for attacks targeting embassy staff or an ambassador.

A compromised site containing the text “Ambassador’s schedule November 2022” was used as a lure to infect visitors with new malware called GraphicalNeutrino.

The threat, which uses the US-based business automation service Notion for command and control (C&C), is a loader that packs numerous anti-analysis capabilities, including sandbox evasion, API unhooking, and string encryption.

According to Recorded Future, which tracks the activity as BlueBravo (PDF), the staging and deployment of the malware is similar to previously observed tactics, techniques, and procedures (TTPs) attributed to APT29.

The lure webpage contained within HTML code an obfuscated ZIP file set to be automatically downloaded on the visitors’ system, showing overlaps with previous observed deployment of the EnvyScout dropper.

The ZIP file contains two DLLs and a benign executable masquerading as a PDF, which was designed to load the libraries using DLL search order hijacking. One of the DLLs contains the GraphicalNeutrino malware, implemented in a thread spawned when the library is initialized.

When launched, GraphicalNeutrino attempts to remove API hooks from specific modules, checks whether persistence is required (which it achieves by creating a new registry key), and then establishes communication with the C&C.

The malware creates a unique identifier for the victim, based on username and computer name, adds the ItIEQ prefix to it, and then uses a Notion API database query filter to determine whether the victim has previously connected to the C&C.

A second, nearly identical GraphicalNeutrino sample that Recorded Future identified and which was compiled only two days after the first, contained only small changes, such as a different Notion database ID, a new identifier prefix, a new key for string decryption, a renamed DLL export function, and modified wait time for C&C communication.

“While we are unable to assess the intended targets of this operation based on the data available, it is likely that ambassadorial or embassy-themed lures are particularly effective during periods of heightened geopolitical tensions, such as is the case with the ongoing war in Ukraine. During such periods, Russian APT groups are highly likely to make extensive use of diplomatically themed lures,” Recorded Future notes.

Related: Analysis of Russian Cyberspy Attacks Leads to Discovery of Windows Vulnerability

Related: Russian Cyberspies Targeting Ukraine Pose as Telecoms Providers

Related: Microsoft Details New Post-Compromise Malware Used by Russian Cyberspies

The post Russia-Linked APT29 Uses New Malware in Embassy Attacks appeared first on SecurityWeek.

US Government Agencies Warn of Malicious Use of Remote Management Software


The Cybersecurity and Infrastructure Security Agency (CISA), National Security Agency (NSA), and Multi-State Information Sharing and Analysis Center (MS-ISAC) are warning organizations of malicious attacks using legitimate remote monitoring and management (RMM) software.

IT service providers use RMM applications to remotely manage their clients’ networks and endpoints, but threat actors are abusing these tools to gain unauthorized access to victim environments and perform nefarious activities.

In malicious campaigns observed in 2022, threat actors sent phishing emails to deploy legitimate RMM software such as ConnectWise Control (previously ScreenConnect) and AnyDesk on victims’ systems, and abuse these for financial gain.

The observed attacks focused on stealing money from bank accounts, but CISA, NSA, and MS-ISAC warn that the attackers could abuse RMM tools as backdoors to victim networks and could sell the obtained persistent access to other cybercriminals or to advanced persistent threat (APT) actors.

Last year, multiple federal civilian executive branch (FCEB) employees were targeted with help desk-themed phishing emails, both via personal and government email addresses.

Links included in these messages directed the victims to a first-stage malicious domain, which automatically triggered the download of an executable designed to connect to a second-stage domain and download RMM software from it, as portable executables that would connect to attacker-controlled servers.

“Using portable executables of RMM software provides a way for actors to establish local user access without the need for administrative privilege and full software installation—effectively bypassing common software controls and risk management assumptions,” the US government agencies warn.

In some cases, the email’s recipient was prompted to call the attackers, who then attempted to convince them to visit the malicious domain.

In October 2022, Silent Push uncovered similar malicious typosquatting activity, in which the adversaries impersonated brands such as Amazon, Geek Squad, McAfee, Microsoft, Norton, and PayPal to distribute the remote monitoring tool WinDesk.Client.exe.

In the attacks targeting federal agencies, the threat actors used the RMM tools to connect to the recipient’s system, then entice them to log into their bank account.

The attackers used the unauthorized access to modify the victim’s bank account summary to show that a large amount of money had been mistakenly refunded, instructing the individual to send the amount back to the scam operator.

“Although this specific activity appears to be financially motivated and targets individuals, the access could lead to additional malicious activity against the recipient’s organization—from both other cybercriminals and APT actors,” CISA, NSA, and MS-ISAC note.

The agencies underline that any legitimate RMM software could be abused for nefarious purposes, that the use of portable executables allows attackers to bypass existing policies and protections, that antivirus defenses would not be typically triggered by legitimate software, and that RMM tools provide attackers with persistent backdoor access to an environment, without the use of custom malware.

CISA, NSA, and MS-ISAC also warn that the legitimate users of RMM software, such as managed service providers (MSPs) and IT help desks, are often targeted by cybercriminals looking to gain access to a large number of the victim MSP’s customers, which could lead to cyberespionage or to the deployment of ransomware and other types of malware.

To stay protected, organizations are advised to implement phishing protections, audit remote access tools, review logs to identify the abnormal use of RMM software, use security software to detect the in-memory execution of RMM software, implementing proper application control policies, restrict the use of RMM software from within the local network, and train employees on phishing.

Related: CISA Updates Infrastructure Resilience Planning Framework

Related: NSA, CISA Explain How Threat Actors Plan and Execute Attacks on ICS/OT

Related: NSA Publishes Best Practices for Improving Network Defenses

The post US Government Agencies Warn of Malicious Use of Remote Management Software appeared first on SecurityWeek.

Chinese Hackers Adopting Open Source ‘SparkRAT’ Tool


A Chinese threat actor tracked as DragonSpark has been using the SparkRAT open source remote administration tool (RAT) in recent attacks targeting East Asian organizations, cybersecurity firm SentinelOne reports.

Relatively new, SparkRAT is a multi-platform RAT written in Golang that can run on Windows, Linux, and macOS systems, and which can update itself with new versions available through its command and control (C&C) server.

The threat uses the WebSocket protocol to communicate with the C&C server and includes support for over 20 commands that allow it to execute tasks, control the infected machine, manipulate processes and files, and steal various types of information.

The malware appears to be used by multiple adversaries but, according to SentinelOne, DragonSpark represents the first cluster of activity where SparkRAT has been constantly deployed in attacks.

The attackers were also seen using the China Chopper webshell, along with other malware tools created by Chinese developers, including BadPotato, GotoHTTP, SharpToken, and XZB-1248, as well as two custom malware families, ShellCode_Loader and m6699.exe.

The m6699.exe malware uses Golang source code interpretation to evade detection, where the Yaegi framework is used “to interpret at runtime encoded Golang source code stored within the compiled binary, executing the code as if compiled”, SentinelOne says.

DragonSpark was seen targeting web servers and MySQL database servers for initial compromise and then performing lateral movement, escalating privileges, and deploying additional malware hosted on attacker-controlled infrastructure.

The cybersecurity firm has observed DragonSpark abusing compromised infrastructure of legitimate organizations in Taiwan, including an art gallery, a baby products retailer, and games and gambling websites, for malware staging.

DragonSpark also uses malware staging infrastructure in China, Hong Kong, and Singapore, while its C&C servers are located in Hong Kong and the US.

Based on the infrastructure and tools, SentinelOne assesses that DragonSpark is a Chinese-speaking adversary, focused either on espionage or cybercrime – one of their C&C IPs was previously linked to the Zegost malware, an information stealer used by Chinese threat actors.

“The threat actor behind DragonSpark used the China Chopper webshell to deploy malware. China Chopper has historically been consistently used by Chinese cybercriminals and espionage groups […]. Further, all of the open source tools used by the threat actor conducting DragonSpark attacks are developed by Chinese-speaking developers,” SentinelOne notes.

Related: Chinese Hackers Exploited Fortinet VPN Vulnerability as Zero-Day

Related: Chinese Cyberspies Targeted Japanese Political Entities Ahead of Elections

Related: Self-Replicating Malware Used by Chinese Cyberspies Spreads via USB Drives

The post Chinese Hackers Adopting Open Source ‘SparkRAT’ Tool appeared first on SecurityWeek.

Microsoft Office to Block XLL Add-ins From Internet


Microsoft is getting ready to improve the protection of Office users by automatically blocking more content sourced from the internet.

Building on previous restrictions that applied to macros in Word and Excel documents, the company is now preparing to block XLL add-ins in Excel files.

XLL add-ins are dynamic link library (DLL) files written in C or C++, and which can only be opened in Excel.

Over the past several years, threat actors have been abusing XLL files for the distribution of malware, typically in phishing campaigns that either deliver the XLL as an attachment, or direct the intended victims to malicious websites from where the XLL is automatically downloaded.

“In order to combat the increasing number of malware attacks in recent months, we are implementing measures that will block XLL add-ins coming from the internet,” the latest entry in the Microsoft 365 roadmap reads.

For the time being, the feature is only in development, with intended worldwide general availability set for March 2023.

The blocking of XLL add-ins is the latest step Microsoft is taking towards preventing the use of malicious Office documents for the delivery of malware and for other malicious purposes.

For years, Office documents downloaded from the internet have been automatically opened in Protected View, with a yellow notification being displayed at the top of the document warning users not to trust internet-sourced files.

However, an ‘Enable editing’ button on the notification allows users to exit Protected View and edit the document’s content, but which also results in any macro code included in the file being automatically executed.

To further strengthen the security of its users, Microsoft last year announced that the yellow notification for documents coming from unknown or untrusted sources is being replaced with a red warning that does not allow users to enable macros with a single click. The company also started restricting all Excel 4.0 (XLM) macros by default.

The post Microsoft Office to Block XLL Add-ins From Internet appeared first on SecurityWeek.